Cranfield Online Research Data (CORD)
Browse
Code_DeepOnet.zip (15.26 MB)

Prediction of Flight Delay using Deep Operator Network with Gradient-mayfly Optimisation Algorithm

Download (15.26 MB)
software
posted on 2024-02-01, 08:47 authored by Desmond Bala BisanduDesmond Bala Bisandu, Irene MoulitsasIrene Moulitsas

Data: 

This folder contains:

- Datasets called Jan_2021_ontime.csv and Nov_2021_ontime.csv were used to obtain the results presented in the journal paper.


Source code:

This folder contains two files having the instructions on how to run the code and a list of library requirements and folders for each of the ML models as named exactly as contained in the paper which implements the proposed Deep Operator Network with Gradient-mayfly Optimisation Algorithm and all the algorithms presented and validated in the journal paper.



Output:

This folder contains:

- Figures called Figure_1_MAE.png, Figure_1_MAPE.png, Figure_1_RMSE.png, Figure_1_MSE.png, Figure_2_MAE.png,Figure_2_MAPE.png, Figure_2_RMSE.png, Figure_2_MSE.png which shows results from the models based on different train/test ratios, The models are: (A)DBN, (B) Gradient Boosting Classifier, (C) Information Gain-SVM, (D) Multi-Agent Approach, (E) DeepLSTM, (F) SSDCA-based Deep LSTM, (G) DeepONet and (H) Proposed GMOA-based DeepOnet.- Figures called Figure 6A.jpg and Figure 6B.jpg which show the EEG signals before applying the preprocessing pipeline, and after applying the preprocessing pipeline, respectively.

- Figures called Figure_1_Prediction_Result_Jan_2021.png and Figure_2_Prediction_Result_Nov_2021.png which are the plots of the prediction results from presented in the journal paper.

- 8 csv files called 1 MAE.csv, 1 MAPE.csv, 1 RMSE.csv, 1 MSE.csv, 2 MAE.csv, 2 MAPE.csv, 2 RMSE.csv, 2 MSE.csv, which contains the evaluation results produced by all algorithms presented and validated in the journal paper.

- 2 csv files called Delay Prediction_Jan_2021_ontime_Figure_1 and Delay Prediction_Nov_2021_ontime_Figure_2 which contains the prediction results produced by all algorithms presented and validated in the journal paper.


Funding

UK Research and Innovation

Petroleum Technology Development Fund (PTDF)

History

Authoriser (e.g. PI/supervisor)

i.moulitsas@cranfield.ac.uk